Development of GUI For Automated Oil Palm Tree Counting Based On Remote Sensing Imagery

This paper presents a novel non-destructive approach for individual oil palm tree counting using remote sensing imagery. This study has three main objectives which are to identify the best techniques for tree counting based on SVM model, to assess tree counting techniques for oil palm and to develop graphic user interface (GUI) for oil palm tree counting based on oil palm tree recognition techniques. Ten subset images with size of $400 \times 400$ pixels were used and each dataset are arranged categorically according to type of oil palm, which are matured and young. Three algorithm are used in this study, which are Gray-Level Co-occurrence Matrix (GLCM), Haar and Biorthogonal wavelet and template matching. All three algorithms are tested by using Support Vector Machine (SVM) to see which algorithms gives highest accuracy and will be used for automated oil palm tree counting. Graphical User Interface (GUI) also created for user to use the function in future. Software used in this study is MATLAB and all process of tree counting are carried out in this software.